Lesson 15 ANOVA (analysis of variance)



Similar documents
Lesson 17 Pearson s Correlation Coefficient

I. Chi-squared Distributions

Z-TEST / Z-STATISTIC: used to test hypotheses about. µ when the population standard deviation is unknown

Hypothesis testing. Null and alternative hypotheses

Inference on Proportion. Chapter 8 Tests of Statistical Hypotheses. Sampling Distribution of Sample Proportion. Confidence Interval

1 Computing the Standard Deviation of Sample Means

Example 2 Find the square root of 0. The only square root of 0 is 0 (since 0 is not positive or negative, so those choices don t exist here).

THE ARITHMETIC OF INTEGERS. - multiplication, exponentiation, division, addition, and subtraction

Definition. A variable X that takes on values X 1, X 2, X 3,...X k with respective frequencies f 1, f 2, f 3,...f k has mean

Measures of Spread and Boxplots Discrete Math, Section 9.4

One-sample test of proportions

SECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES

PSYCHOLOGICAL STATISTICS

Incremental calculation of weighted mean and variance

Now here is the important step

The following example will help us understand The Sampling Distribution of the Mean. C1 C2 C3 C4 C5 50 miles 84 miles 38 miles 120 miles 48 miles

THE REGRESSION MODEL IN MATRIX FORM. For simple linear regression, meaning one predictor, the model is. for i = 1, 2, 3,, n

1. C. The formula for the confidence interval for a population mean is: x t, which was

Exam 3. Instructor: Cynthia Rudin TA: Dimitrios Bisias. November 22, 2011

Chapter 7: Confidence Interval and Sample Size

Basic Elements of Arithmetic Sequences and Series


Case Study. Normal and t Distributions. Density Plot. Normal Distributions

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008

CHAPTER 3 THE TIME VALUE OF MONEY

CHAPTER 7: Central Limit Theorem: CLT for Averages (Means)

Chapter 14 Nonparametric Statistics

Practice Problems for Test 3

Properties of MLE: consistency, asymptotic normality. Fisher information.

Mann-Whitney U 2 Sample Test (a.k.a. Wilcoxon Rank Sum Test)

Determining the sample size

University of California, Los Angeles Department of Statistics. Distributions related to the normal distribution

1 Correlation and Regression Analysis

Here are a couple of warnings to my students who may be here to get a copy of what happened on a day that you missed.

GCSE STATISTICS. 4) How to calculate the range: The difference between the biggest number and the smallest number.

, a Wishart distribution with n -1 degrees of freedom and scale matrix.

A Test of Normality. 1 n S 2 3. n 1. Now introduce two new statistics. The sample skewness is defined as:

Math C067 Sampling Distributions

Maximum Likelihood Estimators.

Present Value Factor To bring one dollar in the future back to present, one uses the Present Value Factor (PVF): Concept 9: Present Value

5: Introduction to Estimation

Overview. Learning Objectives. Point Estimate. Estimation. Estimating the Value of a Parameter Using Confidence Intervals

The Stable Marriage Problem

Repeating Decimals are decimal numbers that have number(s) after the decimal point that repeat in a pattern.

5.4 Amortization. Question 1: How do you find the present value of an annuity? Question 2: How is a loan amortized?

Confidence Intervals. CI for a population mean (σ is known and n > 30 or the variable is normally distributed in the.

Confidence Intervals for One Mean

A probabilistic proof of a binomial identity

Chapter 7 - Sampling Distributions. 1 Introduction. What is statistics? It consist of three major areas:

BINOMIAL EXPANSIONS In this section. Some Examples. Obtaining the Coefficients

5 Boolean Decision Trees (February 11)

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method

Descriptive Statistics

CS103X: Discrete Structures Homework 4 Solutions

Solutions to Selected Problems In: Pattern Classification by Duda, Hart, Stork

Unit 8: Inference for Proportions. Chapters 8 & 9 in IPS

Factoring x n 1: cyclotomic and Aurifeuillian polynomials Paul Garrett <garrett@math.umn.edu>

Soving Recurrence Relations

Output Analysis (2, Chapters 10 &11 Law)

LECTURE 13: Cross-validation

How To Solve The Homewor Problem Beautifully

Week 3 Conditional probabilities, Bayes formula, WEEK 3 page 1 Expected value of a random variable

Confidence intervals and hypothesis tests

Statistical inference: example 1. Inferential Statistics

Sampling Distribution And Central Limit Theorem

Modified Line Search Method for Global Optimization

Your organization has a Class B IP address of Before you implement subnetting, the Network ID and Host ID are divided as follows:

Center, Spread, and Shape in Inference: Claims, Caveats, and Insights

Asymptotic Growth of Functions

Sequences and Series

5.3. Generalized Permutations and Combinations

Section 11.3: The Integral Test

Normal Distribution.

Lecture 4: Cauchy sequences, Bolzano-Weierstrass, and the Squeeze theorem

*The most important feature of MRP as compared with ordinary inventory control analysis is its time phasing feature.

Trigonometric Form of a Complex Number. The Complex Plane. axis. ( 2, 1) or 2 i FIGURE The absolute value of the complex number z a bi is

Taking DCOP to the Real World: Efficient Complete Solutions for Distributed Multi-Event Scheduling

Learning objectives. Duc K. Nguyen - Corporate Finance 21/10/2014

Biology 171L Environment and Ecology Lab Lab 2: Descriptive Statistics, Presenting Data and Graphing Relationships

A Recursive Formula for Moments of a Binomial Distribution

Confidence Intervals

PROCEEDINGS OF THE YEREVAN STATE UNIVERSITY AN ALTERNATIVE MODEL FOR BONUS-MALUS SYSTEM

Ranking Irregularities When Evaluating Alternatives by Using Some ELECTRE Methods

hp calculators HP 12C Statistics - average and standard deviation Average and standard deviation concepts HP12C average and standard deviation

(VCP-310)

Chapter 7 Methods of Finding Estimators

Lecture 2: Karger s Min Cut Algorithm

Laws of Exponents Learning Strategies

3. Greatest Common Divisor - Least Common Multiple

COMPARISON OF THE EFFICIENCY OF S-CONTROL CHART AND EWMA-S 2 CONTROL CHART FOR THE CHANGES IN A PROCESS

MEI Structured Mathematics. Module Summary Sheets. Statistics 2 (Version B: reference to new book)

Simple Annuities Present Value.

Chapter 5: Inner Product Spaces

STA 2023 Practice Questions Exam 2 Chapter 7- sec 9.2. Case parameter estimator standard error Estimate of standard error

Cooley-Tukey. Tukey FFT Algorithms. FFT Algorithms. Cooley

Lecture 3. denote the orthogonal complement of S k. Then. 1 x S k. n. 2 x T Ax = ( ) λ x. with x = 1, we have. i = λ k x 2 = λ k.

OMG! Excessive Texting Tied to Risky Teen Behaviors

EGYPTIAN FRACTION EXPANSIONS FOR RATIONAL NUMBERS BETWEEN 0 AND 1 OBTAINED WITH ENGEL SERIES

CHAPTER 11 Financial mathematics

Transcription:

Outlie Variability -betwee group variability -withi group variability -total variability -F-ratio Computatio -sums of squares (betwee/withi/total -degrees of freedom (betwee/withi/total -mea square (betwee/withi -F (ratio of betwee to withi Example Problem Lesso 5 ANOVA (aalysis of variace Note: The formulas detailed here vary a great deal from the text. I suggest usig the otatio I have outlied here sice it will coicide more with what we have already doe, but you might loo at the text versio as well. Use whatever method you fid easiest to uderstad. Variability Please read about this topic o the web page by locatig the ANOVA demostratio or you ca clic here: http://faculty.ucfsu.edu/dwallace/saova.html Betwee group variability ad withi group variability are both compoets of the total variability i the combied distributios. What we are doig whe we compute betwee ad withi variability is to partitio the total variability ito the betwee ad withi compoets. So: Betwee variability withi variability total variability Hypothesis Testig Agai, with ANOVA we are testig hypotheses that ivolve comparisos of two or more populatios. The overall test, however, will idicate a differece betwee ay of the groups. Thus, the test will ot specify which two, or if some or all of the groups differ. Istead, we will coduct a separate test to determie which specific meas differ. Because of this fact the research hypothesis will state simply that at least two of the meas differ. The ull will still state that there are o sigificat differeces betwee ay of the groups (isert as may mu s as you have groups. H 0 : µµµ3 Critical values are foud usig the F-table i your boo. The table is discussed i the example below. Computatio How do we measure variability i a distributio? That is, how do we measure how differet scores are i the distributio from oe aother? You should ow that we use variace as a measure of variability. With ANOVA or aalysis of variace, we compute

a ratio of variaces: betwee to withi variace. Recall that variace is the average square deviatio of scores about the mea. We will compute the same value here, but as the defiitio suggests, it will be called the mea square for the computatios. So, we are computig variace. Recall that whe we compute variace we first fid the sum of the square deviatios, ad the divide by the sample size ( - or degrees of freedom for a sample. s ( X X Sums _ of _ Squares deg rees _ of _ freedom Whe we compute the Mea Square (variace i order to form the F-ratio, we will do the exact same thig: compute the sums of squares ad divide by degrees of freedom. Do t let the formulas itimidate you. Keep i mid that all we are doig is fidig the variace for our betwee factor ad dividig that by the variace for the withi factor. These two variaces will be computig by fidig each sums of squares ad dividig those sums of squares by their respective degrees of freedom. Sums of Squares We will use the same basic formula for sums of squares that we used with variace. While we will oly use the betwee variace ad withi variace to compute the F- ratio, we will still compute the sums of squares total (all values for completeess. Total Sums of Squares X X ( N Note that it is the same formula we have bee usig. The subscript (tot stads for the total. It idicates that you perform the operatio for ALL values i your distributio (all subjects i all groups. Withi Sums of Squares ( X ( X ( X X X X... Notice that each segmet is the same formula for sums of squares we used i the formula for variace ad for the total sums of squares above. What is differet here is that you cosider each group separately. So, the first segmet with the subscript meas you compute the sum of squares for the first group. Group two is labeled with a, but otice that after that we have group istead of a umber. This otatio idicates that you cotiue to fid the sums of squares as you did for the first two groups for however

may groups you have i the problem. So, could be the third group, or if you have four groups the you would do the same sums of squares computatio for the third ad fourth group. Betwee Sums of Squares ( X ( X ( X ( X... N We have the same otatio here. Agai, you perform the same operatio for each separate group i your problem. However, with this formula oce we compute the value for each group we must subtract a operatio at the fial step. This operatio is half the sums of squares we computed for the sums of squares total. Degrees of Freedom Agai, we will first compute the sums of squares for each source of variace, divide the values by degrees of freedom i order to get the two mea square values we eed to form the F-ratio. Degrees of freedom, however, is differet for each source of variability. Total Degrees of freedom N this N value is the total umber of values i all groups Withi Degrees of freedom N K K is the umber of categories or groups, N is still the total N withi Betwee Degrees of freedom Betwee K We will also use degrees of freedom to locate the critical value o the F-table (see page A-9 for alpha.05 ad A-30 for alpha.0. The umerator of the F-ratio is the betwee factor, so we will use the degrees of freedom betwee alog the top of your table. The deomiator of the F-ratio is the withi subjects factor, so will use degrees of freedom withi alog the left margi of the table. Mea Square Now we divide each sums of squares by the respective mea square. Do t let the formula s itimidate you. All we are doig is matchig up degrees of freedom with the Sums of squares to get the mea square (variace Withi Mea Square Betwee Betewee Betwee Betwee Mea Square Withi Withi

F-ratio The fial step is to divide our betwee by withi variace to see if the effect (betwee is large compared to the error (withi. Betwee F Example Withi A therapist wats to examie the effectiveess of 3 therapy techiques o phobias. Subjects are radomly assiged to oe of three treatmet groups. Below are the rated fear of spiders after therapy. Test for a differece at α.05 Therapy A Therapy B Therapy C 5 3 3 0 5 0 4 Σ x 8 Σ x 0 Σ x 3 5 Σ 74 Σ 6 x STEP : State the ull ad alterative hypotheses. H at least oe mea differs H 0 : µ µ µ3 x Σ x 3 7 STEP : Set up the criteria for maig a decisio. That is, fid the critical value. You might do this step after Step 3 sice that is where you compute the critical value. Betwee K 3- withi F critical 3.88 N K 5-3 STEP 3: Compute the appropriate test-statistic. Although i this example I have give the summary values, for some problems you might have to compute the sum of x, ad sum of squared x s yourself. ( X X ( 33 07 N 5 089 07 5 07 7.6 34.4

( X ( X ( X X X X... ( 8 ( 0 ( 5 74 6 7 5 5 5 34 00 5 74 6 7 5 5 5 ( 74 64.8 ( 6 0 ( 7 5 9. 6 7. ( X ( X ( X ( X ( 8 ( 0 ( 5 ( 33 5 5 5... N 5 34 00 5 089 5 5 5 5 64.8 0 5 7.6 7. Note that aytime you compute two of the Sums of Squares you ca derive the third oe without computatio because Betwee Withi Total tot N Betwee K withi N K tot 5 4 3 withi 5 3 Betwee Betwee Withi Betewee F Betwee Withi 7. 7. 8.6 Betewee 8.6. 43 F 6.43 Betwee Oce we have computed all the values, very ofte we place them i a source table (below. Puttig the values i a table lie this oe may mae it easier to thi about the statistic. Notice that oce we get the Sums of Squares o the table, we will divide those values by the i the ext colum. Oce we get the two mea squares we divide those to get F. Withi

Source F Betwee (group 7. 8.6 6_ Withi (error _7..43_ Total _34.4 4 STEP 4: Evaluate the ull hypothesis (based o your aswers to the above steps. Reject the ull STEP 5: Based o your evaluatio of the ull hypothesis, what is your coclusio? There is at least oe group that is differet from at least oe other group.